One Model to Serve All: Star Topology Adaptive Recommender for Multi-Domain CTR Prediction

Abstract

Traditional industrial recommenders are usually trained on a single business domain and then serve for this domain. However, in large commercial platforms, it is often the case that the recommenders need to make click-through rate (CTR) predictions for multiple business domains. Different domains have overlapping user groups and items. Thus, there exist commonalities. Since the specific user groups have disparity and the user behaviors may change in various business domains, there also have distinctions. The distinctions result in domain-specific data distributions, making it hard for a single shared model to work well on all domains. To learn an effective and efficient CTR model to handle multiple domains simultaneously, we present Star Topology Adaptive Recommender (STAR). Concretely, STAR has the star topology, which consists of the shared centered parameters and domain-specific parameters. The shared parameters are applied to learn commonalities of all domains, and the domain-specific parameters capture domain distinction for more refined prediction. Given requests from different business domains, STAR can adapt its parameters conditioned on the domain characteristics. The experimental result from production data validates the superiority of the proposed STAR model. Since 2020, STAR has been deployed in the display advertising system of Alibaba, obtaining averaging 8.0% improvement on CTR and 6.0% on RPM (Revenue Per Mille).

arXiv commentAccepted at CIKM 2021

References in corpus (9)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• An Overview of Multi-Task Learning in Deep Neural Networks
• On Calibration of Modern Neural Networks
• xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems
• Layer Normalization
• Multi-Task Learning as Multi-Objective Optimization
• Multi-Interest Network with Dynamic Routing for Recommendation at Tmall
• Multi-Domain Adversarial Learning
• One Person, One Model, One World: Learning Continual User Representation without Forgetting

Cited by in corpus (7)

• HAMUR: Hyper Adapter for Multi-Domain Recommendation
• Automatic Expert Selection for Multi-Scenario and Multi-Task Search
• Exploring Adapter-based Transfer Learning for Recommender Systems: Empirical Studies and Practical Insights
• An Unified Search and Recommendation Foundation Model for Cold-Start Scenario
• Scenario-Adaptive and Self-Supervised Model for Multi-Scenario Personalized Recommendation
• Multi-Scenario Ranking with Adaptive Feature Learning
• Real Negatives Matter: Continuous Training with Real Negatives for Delayed Feedback Modeling